JPH07168586A - Voice detecting device - Google Patents

Abstract

PURPOSE:To prevent degradation in processing performance which eliminates degradation in voice quality and a noise and to always listen a voice in the best state. CONSTITUTION:This device is provided with analog/digital conversion processing sections 4f, 4s which convert detected signals from microphones 2, 3 to a digital first signal Df and a digital second signal Ds, a noise cancel functioning section 5 which outputs a digital output signal Do in which the second signal Ds is filtered by an adaptive filter, an output signal of the adaptive filter is subtracted from the first signal Df, and a noise is canceled, and signal level adjusting sections 7f, 7s which adjust levels of input signals Sif, Sis inputted to the analog/ digital conversion processing sections 4f, 4s so as to be made constant corresponding to a timing signal St based on process contents of the noise cancel functioning section 5.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice detecting apparatus having a noise canceling function section using an adaptive filter.

[0002]

2. Description of the Related Art Generally, a voice detecting device having a noise canceling function is used for detecting voice in a noisy environment.

FIG. 5 shows a conventional voice detection device 50. The voice detection device 50 includes a first detection unit 51 including a microphone for detecting voice and noise and an analog-digital conversion processing unit, a second detection unit 52 including a microphone for detecting noise and an analog-digital conversion processing unit, and an adaptive filter. (Adaptive filter) 53 and subtraction processing unit 5
4 is provided and configured. Such a voice detection device 50 is
The first digital signal Da is obtained from the first detector 51, and the second digital signal Db is obtained from the second detector 52. Then, the digital second signal Db is filtered by the adaptive filter 53, and the subtraction processing unit 54 subtracts the output signal Dc of the adaptive filter 53 from the digital first signal Da, whereby the output of the subtraction processing unit 54 is obtained. A voice signal containing only the voice component with the noise component canceled is obtained.

However, such a conventional voice detecting device 5 is used.
When the noise level is 0, the level of the detection signal output from the microphone may become unstable when the noise level of the usage environment is high or due to the characteristics of the connected microphone. Therefore, for example, when the signal level becomes excessive, distortion due to saturation of the signal level causes deterioration of voice quality, and when the signal level becomes too small, noise is removed. There is a problem that the conventional voice detection device 50 cannot sufficiently achieve the original purpose of making it easier to hear the voice, such as a decrease in processing performance.

The present invention solves the problems existing in the prior art as described above, and prevents the deterioration of the voice quality and the processing performance of removing noise, so that the voice can always be heard in the best condition. An object is to provide a voice detection device.

[0006]

According to the present invention, a first microphone 2 capable of detecting voice and noise, a second microphone 3 capable of detecting noise, and each microphone 2, 3 are provided.
The analog-to-digital conversion processing units 4f and 4s for converting the detection signals from the above into the digital first signal Df and the digital second signal Ds respectively, and the digital second signal Ds is filtered by the adaptive filter, and from the digital first signal Df When configuring the voice detection device 1 including the noise canceling function unit 5 that outputs the digital output signal Do in which noise is canceled by subtracting the output signal of the adaptive filter, particularly based on the processing content of the noise canceling function unit 5. Timing signal S
corresponding to t, the analog-digital conversion processing unit 4f,
It is characterized in that signal level adjusting sections 7f and 7s for adjusting the levels of the input signals Sif and Sis input to the 4s to be constant are provided.

In this case, the signal level adjusters 7f and 7s make adjustments in response to the timing signal St based on the period during which the adaptive filters 8a ... Do not execute the filter coefficient changing process. The signal level adjusting unit 7s (same for 7f) is a variable gain amplifier 9 that amplifies the detection signal from the second microphone 3 and an input signal that is output from the variable gain amplifier 9 and input to the analog-digital conversion processing unit 4s. A signal level detector 10 for detecting the level of Sis, and a signal level detector 10 corresponding to the timing signal St
The gain control unit 11 can be configured to vary the gain of the variable gain amplifier 9 so that the level of the input signal Sis becomes constant based on the detection result. Furthermore, the gain control unit 1
In No. 1, the gain of the variable gain amplifier 9 can be changed stepwise by a preset constant level adjustment amount.

The signal level adjusters 7f and 7s may be provided on both the first microphone 2 side and the second microphone 3 side, or may be provided on either one, especially only on the second microphone 3 side. .

Further, the voice detection device 1 has a digital filter section 12 for dividing the digital first signal Df and the digital second signal Ds into a plurality of bands, and a plurality of noise canceling function sections 5a, 5b corresponding to the respective bands. 5c
, And the addition processing unit 13 for adding the digital output signals Doa, Dob, and Doc of the noise canceling function units 5a ...
It is desirable to configure with.

[0010]

According to the voice detection device 1 of the present invention, the first microphone 2 detects the voice including the noise and the second microphone 3 detects the noise, and the detection signals from the respective microphones 2 and 3 are It is given to the signal level adjusting units 7f and 7s, respectively. In the signal level adjusting units 7f and 7s, the levels of the input signals Sif and Sis input to the analog-digital conversion processing units 4f and 4s become constant in correspondence with the timing signal St based on the processing content of the noise canceling function unit 5. Is adjusted. Then, the adjusted input signals Sif and Sis are converted into the digital first signal Df and the digital second signal Ds by the analog-digital conversion processing units 4f and 4s, respectively.

At this time, in the signal level adjusting section 7s (same for 7f), the variable gain amplifier 9 amplifies the detection signal from the second microphone 3, and the signal level detecting section 10 outputs the variable gain amplifier 9 from the variable gain amplifier 9. The level of the input signal Sis that is output and input to the analog-digital conversion processing unit 4s is detected. Further, since the detection result of the signal level detection unit 10 is given to the gain control unit 11, the gain control unit 11 inputs the input signal Sis input to the analog-digital conversion processing unit 4s based on the detection result.
The gain of the variable gain amplifier 9 is changed so that the level of is constant.

Specifically, when the level of the detection signal from the second microphone 3 becomes high (low), the analog-
Since the level of the input signal Sis input to the digital conversion processing unit 4s also becomes high (low), the signal level detection unit 10 detects this signal level. On the other hand, since this detection result is given to the gain control unit 11, the gain control unit 11 variably controls the gain of the variable gain amplifier 9 to be low (high). As a result, the signal level output from the variable gain amplifier 9 becomes low (high), and the level of the input signal Sis input to the analog-digital conversion processing unit 4s is maintained constant.

The adjustment by the signal level adjusting unit 7s (7f is also the same) is performed at the timing corresponding to the timing signal St given. This timing is a period during which the filter coefficient changing process in the adaptive filters 8a ... Is not executed, whereby the influence of the change in the signal level on the processing content of the noise canceling function section 5 is reduced. Furthermore, the gain control unit 11 changes the gain of the variable gain amplifier 9 stepwise by a preset constant level adjustment amount, thereby preventing a sudden level change.

On the other hand, the digital first signal Df and the digital second signal Ds are processed by the digital filter unit 12.
Each is divided into a plurality of bands. Then, the noise canceling processing is performed by each of the plurality of noise canceling function units 5a, 5b, 5c corresponding to each band. That is, in one noise canceling function section 5a (5b and 5c are the same), the digital second signal Dsa is filtered by the adaptive filter 8a, and at the same time the digital first signal Dfa is changed to the adaptive filter 8a.
Output signal Dma is subtracted. As a result, the noise cancel function units 5a, 5b, and 5c output digital output signals Doa and D with noise canceled for each band.
ob and Doc are obtained. Then, each digital output signal D
oa, Dob, and Doc are added by the addition processing unit 13, and the digital output signal Do output from the addition processing unit 13 is converted into an analog signal Ao by the digital-analog conversion processing unit 6, for example. Therefore, the analog signal Ao becomes a noise-free audio signal. When performing digital recording or the like, the digital output signal Do output from the addition processing unit 13 is directly used.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, preferred embodiments according to the present invention will be described in detail with reference to the drawings.

First, the configuration of the voice detection device 1 according to this embodiment will be described.

FIG. 1 shows a block circuit of a voice detecting device 1, 2 is a first microphone capable of detecting voice and noise, and 3 is a second microphone capable of detecting noise. As shown in FIG. 4, the microphones 2 and 3 are integrally provided in a single casing 21. in this case,
Each of the microphones 2 and 3 uses a directional microphone, and the sound receiving unit 2 of each of the microphones 2 and 3 is used.
f and 3f are arranged so that they face each other by 180 °.
The microphones 2 and 3 may be arranged separately.

Reference numeral 22 denotes an opposite phase (or in-phase) in each detection signal from the first microphone 2 and the second microphone 3.
Is a homophonic signal component removing circuit for removing the signal component of.

Reference numerals 7f and 7s are signal level adjusting units for adjusting the levels of the detection signal from the first microphone 2 and the detection signal from the second microphone 3 output from the homophonic signal component removing circuit 22, respectively. FIG. 2 shows the signal level adjusting unit 7s in a concrete form. In addition, in FIG.
Although one signal level adjusting unit 7f is omitted, it can be configured similarly to the signal level adjusting unit 7s. In the figure, 9
Is a variable gain amplifier for amplifying the detection signal given from the second microphone 3, and 10 is a signal level for detecting the level of the input signal Sis output from the variable gain amplifier 9 and input to the analog-digital conversion processing unit 4s described later. The detection unit 11 is a gain control unit that responds to the timing signal St and changes the gain of the variable gain amplifier 9 so that the level of the input signal Sis becomes constant based on the detection result of the signal level detection unit 10. . In this case, the gain control unit 11
Is a signal for enabling adjustment by the signal level adjusting unit 7s during the period when the filter coefficient changing process in the adaptive filter 8a ... Is not executed, and is given by the state signal generating unit 26 described later. To be done.

Reference numerals 4f and 4s denote analog-to-digital conversion processing units, which are input signals Sif and Sis input from the signal level adjusting units 7f and 7s (gain variable amplifier 9 ...) to a first digital signal Df and a second digital signal. Convert to Ds respectively.

Reference numeral 12 is a digital filter section for dividing the digital first signal Df and the digital second signal Ds into a plurality of bands. In the embodiment, the divided bands are set to the low band, the middle band, and the high band, respectively. Therefore, the digital filter unit 12 filters a pair of low-pass filters 23a and 24a having the same characteristic for filtering the low band, a pair of mid-band filters 23b and 24b having the same characteristic for filtering the middle band, and a high band. It is composed of a pair of high-pass filters 23c and 24c having the same characteristics. In this case, the filters 23a ..., 24a
It is desirable to use a digital FIR filter having a linear phase characteristic for ..., By this, phase shift in each band can be prevented.

Reference numeral 5 denotes a noise canceling function section, which is composed of three noise canceling function sections 5a, 5b and 5c corresponding to respective bands, that is, low band, middle band and high band. The noise canceling function unit 5a for processing the low frequency band includes an adaptive filter 8a for filtering the digital second signal Dsa obtained from the low frequency band filter 24a, an output signal Dma of the adaptive filter 8a and a digital first signal Df.
It is configured by including a subtraction processing unit 25a that performs a subtraction process on the output signal Dfa from the low-pass filter 23a on the side. On the other hand, noise canceling function units 5b and 5c for processing the mid range and high range
Is also configured similarly to the noise canceling function unit 5a.
8b and 8c are adaptive filters, and 25b and 2c.
Reference numeral 5c indicates a subtraction processing unit. Also, each of the adaptive filters 8a, 8b, 8c is connected to the state signal generator 26.

An addition processing unit 13 has a function of adding the digital output signals Doa, Dob, Doc output from the noise canceling function units 5a, 5b, 5c. A digital-analog conversion processing unit 6 converts the digital output signal Do output from the addition processing unit 13 into an analog signal Ao.

Next, the operation of the voice detection device 1 according to this embodiment will be described.

First, when voice is not generated, the filter coefficients of the adaptive filters 8a, 8b, 8c provided in the noise canceling function units 5a, 5b, 5c are set (changed). In this case, the voice including noise is detected by the first microphone 2 and the noise is detected by the second microphone 3. The detection signal (analog signal) from each of the microphones 2 and 3 is given to the same-sound-signal component removing circuit 22, and the opposite phase (or the same phase) in each detection signal is given.
Signal components, for example, signal components such as a collision sound of an object hitting the casing 21 are removed. Then, the detection signals from the microphones 2 and 3 output from the homologous signal component removal circuit 22 are signal level adjusting units 7f and 7s, respectively.
And is adjusted to be constant at the timing corresponding to the timing signal St based on the processing content of the noise canceling function unit 5.

That is, in the signal level adjusting unit 7s (same for 7f), the variable gain amplifier 9 amplifies the detection signal from the second microphone 3, and the signal level detecting unit 10 outputs the variable gain amplifier 9 from the variable gain amplifier 9. Input signal Si to be input to the analog-digital conversion processing unit 4s
The level of s is detected. In addition, the signal level detection unit 10
Since the detection result obtained by is given to the gain control unit 11, the gain control unit 11 controls the gain variable amplifier so that the level of the input signal Sis input to the analog-digital conversion processing unit 4s becomes constant based on the detection result. The gain of 9 is variably controlled.

In this case, as shown in FIG. 3A, when the level of the detection signal Sds from the second microphone 3 becomes high, the level of the input signal Sis input to the analog-digital conversion processing section 4s also becomes high. Therefore, this signal level is detected by the signal level detection unit 10. Then, this detection result is given to the gain control unit 11, and the gain control unit 11 variably controls the gain of the variable gain amplifier 9 to be low. As a result, the signal level output from the variable gain amplifier 9 becomes low, and the level of the input signal Sis input to the analog-digital conversion processing unit 4s is as shown in FIG.
As shown in (c), it is maintained at a constant level −Lr to Lr. On the other hand, as shown in FIG. 3B, when the level of the detection signal Sds from the second microphone 3 becomes low,
Since the level of the input signal Sis input to the analog-digital conversion processing unit 4s also becomes low, this signal level is detected by the signal level detection unit 10. Then, this detection result is given to the gain control section 11, and the gain control section 11 variably controls the gain of the variable gain amplifier 9 so as to be high. As a result, the signal level output from the variable gain amplifier 9 becomes high, and the analog-digital conversion processing unit 4s
The level of the input signal Sis input to is maintained at a constant level −Lr to Lr as shown in FIG.

At this time, the signal level adjusting unit 7s (7
The adjustment by (f is also the same) is controlled so as to be performed at a timing corresponding to the timing signal St given from the state signal generating unit 26, that is, in a period in which the filter coefficient changing process in the adaptive filters 8a ... Is not executed. As a result, the influence of the change in the signal level on the processing content of the noise canceling function unit 5 is reduced. Furthermore, the gain control unit 11 can vary the gain of the variable gain amplifier 9 stepwise by a preset constant level adjustment amount (for example, about 0.5 dB). This prevents a sudden level change.

On the other hand, the input signal S whose signal level is adjusted
if and Sis are analog-digital conversion processing units 4
It is added to f and 4s and converted into a digital first signal Df and a digital second signal Ds, respectively. Further, the digital first signal Df and the digital second signal Ds are divided by the digital filter unit 12 into three bands of low band, middle band and high band, respectively. That is, the low-pass filter 23a,
A digital first signal Dfa and a digital second signal Dsa containing only low-frequency components are obtained from the digital first signal Df and the digital second signal Ds provided to 24a, and are provided to the mid-range filters 23b and 24b. From the digital first signal Df and the digital second signal Ds, a digital first signal Dfb and a digital second signal Dsb containing only the mid-range component are obtained, respectively, and further, the digital first signals Dfb to the high-pass filters 23c and 24c are provided. From the first signal Df and the second digital signal Ds, the first digital signal Dfc and the second digital signal Dfc containing only high frequency components
get sc respectively.

Then, the filter coefficients of the adaptive filters 8a, 8b, 8c are changed by the noise canceling function units 5a, 5b, 5c corresponding to the respective bands. That is, in the low-frequency noise cancellation function unit 5a, the digital second signal Dsa is filtered by the adaptive filter 8a, and further, the subtraction processing unit 25a subtracts the output signal Dma of the adaptive filter 8a from the digital first signal Dfa. The signal Dea is obtained, and the filter coefficient of the adaptive filter 8a is changed so that the differential signal Dea becomes zero. At this time, the state signal Sc for changing the filter coefficient is given from the state signal generator 26 to the adaptive filter 8a. On the other hand, also in the noise canceling function units 5b and 5c, the filter coefficient is changed similarly to the noise canceling function unit 5a. In addition, Dm
a and Dmc are output signals of the adaptive filters 8b and 8c, and Deb and Dec are difference signals obtained from the subtraction processing units 25b and 25c.

On the other hand, at the time of actual voice detection, the status signal S for fixing the filter coefficient is supplied from the status signal generator 26.
s is given to each adaptive filter 8a, 8b, 8c. In this case, the filter coefficient is fixed by the state signal Ss, but in addition, the change width (step gain) of the filter coefficient may be made smaller than the change width by the state signal Sc by the state signal Ss. Therefore, the changing method of the filter coefficient can be changed by the state signals Sc and Ss.

Then, by similarly picking up voice and noise from the microphones 2 and 3, the subtraction processing units 25a, 25b, and 25c receive the digital first signal Dfa,
From Dfb and Dfc to adaptive filters 8a and 8b,
8c output signals Dma, Dmb, Dmc are subtracted to cancel noise-free digital output signals Doa, Do
b and Doc are output. Then, the respective digital output signals Doa, Dob, Doc are added by the addition processing unit 13 and output as the digital output signal Do,
The digital output signal Do is converted into an analog signal Ao by the digital-analog conversion processing unit 6, and a target voice signal with noise canceled is obtained.

As described above, in the embodiment, since the first digital signal including the voice and the noise and the second digital signal including the noise are divided into a plurality of bands to be processed, the noise including a wide dynamic range is included. However, there is an advantage that noise can be sufficiently removed and high-quality speech can be detected.

The embodiment has been described in detail above.
The present invention is not limited to such an embodiment.
For example, voice is a concept including sounds for all purposes to be detected. If necessary, a bandpass filter (digital FIR filter) may be inserted before the digital-analog conversion processing section. On the other hand, the signal level adjusting unit may be provided only on the microphone side that detects noise. It is not always necessary to provide the homologous signal component removing circuit. Furthermore, as a "timing signal based on the processing content of the noise cancellation function unit",
Although the "timing signal based on the period during which the adaptive filter does not execute the process of changing the filter coefficient" is illustrated,
It is not limited to this processing content. For example, when the filter coefficient changing process is executed in a state where the change amount is small, the level adjustment in the present invention can be performed at the same time. Also, the filter coefficient changing process and the level adjustment may be performed simultaneously immediately after the power is turned on. In addition, the detailed configuration, method, and the like can be arbitrarily changed without departing from the scope of the present invention.

[0035]

As described above, according to the present invention, the first microphone capable of detecting voice and noise, the second microphone capable of detecting noise, and the detection signals from the respective microphones are the digital first signal and the digital second signal, respectively. An analog-to-digital conversion processing unit for converting into a signal, and a noise canceller that filters a digital second signal with an adaptive filter and subtracts the output signal of the adaptive filter from the digital first signal to cancel the noise to output a digital output signal In a voice detection device including a functional unit, a signal level adjusted so that the level of an input signal input to the analog-digital conversion processing unit is constant, corresponding to a timing signal based on the processing content of the noise canceling functional unit. Since the adjustment unit is provided, the voice It is possible to prevent deterioration of the performance of removing reduced and noise quality, always exhibits a remarkable effect that a voice can be heard in the best condition.

[Brief description of drawings]

FIG. 1 is a block circuit diagram of a voice detection device according to the present invention,

FIG. 2 is a block circuit diagram of a signal level adjusting unit in the voice detection device,

FIG. 3 is a signal waveform diagram for explaining a function of a signal level adjusting unit in the voice detection device,

FIG. 4 is a layout view of microphones in the voice detection device,

FIG. 5 is a block circuit diagram of a voice detection device according to the related art;

[Explanation of symbols]

 1 Voice Detection Device 2 First Microphone 3 Second Microphone 4f Analog-Digital Conversion Processing Unit 4s Analog-Digital Conversion Processing Unit 5 Noise Canceling Function Unit 5a ... Noise Canceling Function Unit 7f Signal Level Adjusting Unit 7s Signal Level Adjusting Unit 8a ... Adaptive Filter 9 Variable gain amplifier 10 Signal level detection unit 11 Gain control unit 12 Digital filter unit 13 Addition processing unit Df Digital first signal Ds Digital second signal Do Digital output signal Doa ... Digital output signal St Timing signal Sif Input signal Sis input signal

Claims (5)

[Claims] 1. A first microphone capable of detecting voice and noise, a second microphone capable of detecting noise, and an analog-digital converter for converting detection signals from the respective microphones into a digital first signal and a digital second signal, respectively. A voice that includes a conversion processing unit and a noise canceling function unit that filters the digital second signal with an adaptive filter and subtracts the output signal of the adaptive filter from the digital first signal to output a digital output signal with noise canceled. In the detection device, a signal level adjusting unit for adjusting the level of the input signal input to the analog-digital conversion processing unit to be constant is provided in accordance with the timing signal based on the processing content of the noise canceling function unit. Characteristic voice detection device. 2. The voice detecting apparatus according to claim 1, wherein the signal level adjusting unit adjusts in response to a timing signal based on a period during which the adaptive filter does not perform a filter coefficient changing process. 3. A signal level adjusting unit, a variable gain amplifier for amplifying a detection signal applied from the first microphone and / or the second microphone, and an output from the variable gain amplifier for input to an analog-digital conversion processing unit. A signal level detection unit that detects the level of the input signal, and a gain that corresponds to the timing signal and that makes the level of the input signal that is input to the analog-digital conversion processing unit constant based on the detection result of the signal level detection unit. 3. The voice detection device according to claim 1, further comprising a gain control unit that varies the gain of the variable amplifier. 4. The voice detecting apparatus according to claim 3, wherein the gain control section varies the gain of the variable gain amplifier stepwise by a predetermined constant level adjustment amount. 5. A digital filter section for dividing each of the first digital signal and the second digital signal into a plurality of bands, a plurality of noise canceling function sections corresponding to each band, and a digital output signal of each noise canceling function section. The addition processing unit for adding is provided.
The voice detection device described.